Search

Recent growth in population scale sequencing initiatives involve both cohort scale and proportion of genome surveyed, with a transition from genotyping arrays to broader genome sequencing approaches. The resulting datasets can be challenging to analyze. Here we introduce DNARecords a novel sparse-compatible format for large scale genetic data. The structure enables integration of complex data types such as medical images and drug structures towards the development of machine learning methods to predict disease risk and drug response. We demonstrate its speed and memory advantages for various genetics analyses. These performance advantages will become more pronounced as it becomes feasible to analyze variants of lower population allele frequencies. Finally, we provide an open-source software plugin, built on top of Hail, to allow researchers to write and read such records as well as a set of examples for how to use them.

Our respiratory system is vital to protect us from the surrounding nonsterile environment; therefore, it is critical for a state of homeostasis to be maintained through a balance of inflammatory cues. Recent studies have shown that actively transcribed noncoding regions of the genome are emerging as key regulators of biological processes, including inflammation. lincRNA-Cox2 is one such example of an inflammatory inducible long intergenic noncoding RNA functioning to fine-tune immune gene expression. Using bulk and single-cell RNA sequencing, in addition to FACS, we find that lincRNA-Cox2 is most highly expressed in the lung and is most upregulated after LPS-induced lung injury (acute lung injury [ALI]) within alveolar macrophages, where it functions to regulate inflammation. We previously reported that lincRNA-Cox2 functions to regulate its neighboring protein Ptgs2 in cis, and in this study, we use genetic mouse models to confirm its role in regulating gene expression more broadly in trans during ALI. Il6, Ccl3, and Ccl5 are dysregulated in the lincRNA-Cox2–deficient mice and can be rescued to wild type levels by crossing the deficient mice with our newly generated lincRNA-Cox2 transgenic mice, confirming that this gene functions in trans. Many genes are specifically regulated by lincRNA-Cox2 within alveolar macrophages originating from the bone marrow because the phenotype can be reversed by transplantation of wild type bone marrow into the lincRNA-Cox2–deficient mice. In conclusion, we show that lincRNA-Cox2 is a trans-acting long noncoding RNA that functions to regulate immune responses and maintain homeostasis within the lung at baseline and on LPS-induced ALI.

Summary As the use of single-cell technologies has grown, so has the need for tools to explore these large, complicated datasets. The UCSC Cell Browser is a tool that allows scientists to visualize gene expression and metadata annotation distribution throughout a single-cell dataset or multiple datasets. Availability and implementation We provide the UCSC Cell Browser as a free website where scientists can explore a growing collection of single-cell datasets and a freely available python package for scientists to create stable, self-contained visualizations for their own single-cell datasets. Learn more at https://cells.ucsc.edu. Supplementary information Supplementary data are available at Bioinformatics online.

In-silico modeling of patient clinical drug response (CDR) promises to revolutionize personalized cancer treatment. State-of-the-art CDR predictions are usually based on cancer cell line drug perturbation profiles. However, prediction performance is limited due to the inherent differences between cancer cell lines and primary tumors. In addition, current computational models generally do not leverage both chemical information of a drug and a gene expression profile of a patient during training, which could boost prediction performance. Here we develop a Patient Adapted with Chemical Embedding (PACE) dual convergence deep learning framework that a) integrates gene expression along with drug chemical structures, and b) is adapted in an unsupervised fashion by primary tumor gene expression. We show that PACE achieves better discrimination between sensitive and resistant patients compared to the state-of-the-art linear regularized method (9/12 VS 3/12 drugs with available clinical outcomes) and alternative methods.

Deep learning architectures such as variational autoencoders have revolutionized the analysis of transcriptomics data. However, the latent space of these variational autoencoders offers little to no interpretability. To provide further biological insights, we introduce a novel sparse Variational Autoencoder architecture, VEGA (Vae Enhanced by Gene Annotations), whose decoder wiring is inspired by a priori characterized biological abstractions, providing direct interpretability to the latent variables. We demonstrate the interpretability and flexibility of VEGA in diverse biological contexts, by integrating various sources of biological abstractions such as pathways, gene regulatory networks and cell type identities in the latent space of our model. We show that our model could recapitulate the mechanism of cellular-specific response to treatments, the status of master regulators as well as jointly investigate the cell type and cellular state identity in developing cells. We envision the approach could serve as an explanatory biological model in contexts such as development and drug treatment experiments.

SummaryAs the use of single-cell technologies has grown, so has the need for tools to explore these large, complicated datasets. The UCSC Cell Browser is a tool that allows scientists to visualize gene expression and metadata annotation distribution throughout a single-cell dataset or multiple datasets.Availability and implementationWe provide the UCSC Cell Browser as a free website where users can explore a growing collection of single-cell datasets and a freely available python package for scientists to create stable, self-contained visualizations for their own single-cell datasets. Learn more at https://cells.ucsc.edu.Contactcells@ucsc.edu

Pediatric diffuse midline gliomas are lethal cancers, the majority of which harbor the H3 p.K27M mutations. Although it has potential implications on the treatment of diffuse midline glioma as a disease driver; the timing, cell type of origin, and effect of the H3 p.K27M mutation on early embryonic brain development is poorly understood. The purpose of our study is to elucidate the molecular mechanisms by which the histone H3 p.K27M mutation drives tumorigenesis of pediatric diffuse midline gliomas using the analysis of genomic datasets. Here, we performed differential RNA sequencing gene expression analysis of a cohort of H3K27M and H3 wild type (WT) pediatric diffuse midline gliomas, revealing that genes in the epithelial-mesenchymal transition (EMT) pathway were significantly differentially expressed between the mutant and WT tumors. Several EMTs are required for normal brain development. Overall, pre-EMT genes, including the master regulator of EMT SNAI1, were overexpressed in H3K27M tumors compared to the WT tumors, while post-EMT genes were underexpressed. We hypothesized that the H3 p.K27M mutation may lead to gliomagenesis by inducing a stall in the EMT in early brain development. To test this hypothesis, we examined published single-cell RNA sequencing data from pediatric diffuse midline gliomas alongside similar data from organoid models of neural development, collected from multiple developmental timepoints. This analysis revealed transcriptional similarities between H3K27M and pre-EMT neural stem cells. Currently, we are investigating the expression of EMT markers in H3K27M and WT pediatric glioma primary cell lines, using Western blotting, RT-PCR, and CRISPRi screening. In conclusion, we observed aberrant expression of genes involved in EMT in H3K27M pediatric gliomas. Our observations are consistent with a model in which the p.H3K27M mutation is associated with a pre-EMT cell phenotype, potentially due to an arrest in the EMT pathway or de-differentiation of mature astrocytes. Citation Format: ALLISON R. CHENEY, Lauren M. Sanders, Lucas Seninge, Holly C. Beale, Ellen Towle Kephart, Jacob Pfeil, Katrina Learned, A. Geoffrey Lyle, Isabel Bjork, David Haussler, Sofie R. Salama, Olena M. Vaske. H3K27M gliomas are characterized by a stall in the epithelial-mesenchymal transition [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 6154.

The purpose of our study was to elucidate the molecular mechanisms by which the histone H3 K27M mutation drives tumorigenesis of pediatric gliomas. Though it has potential implications on the treatment of diffuse midline glioma as a disease driver, the timing, cell type of origin, and effect of the H3K27M mutation on early embryonic brain development are not fully characterized. Here, we performed differential expression analysis on a cohort of H3K27M and H3WT pediatric gliomas, revealing that genes in the epithelial-mesenchymal transition (EMT) pathway were significantly differentially expressed. SNAI1, the EMT master regulator, was significantly overexpressed in the H3K27M tumor cohort. Overall, pre-EMT genes were overexpressed in H3K27M tumors while post-EMT genes were underexpressed. We hypothesized that H3K27M may lead to gliomagenesis by stalling an EMT in early brain development and employed single-cell and bulk RNA sequencing data from cerebral organoids at multiple developmental timepoints to test this hypothesis. We observed that a long noncoding RNA (lncRNA) signature identified as transiently expressed in early brain development was preferentially expressed in H3K27M tumors. Cell type-specific lncRNA signatures had higher expression in H3K27M tumors for pre-EMT cell types, and higher expression in H3WT tumors for post-EMT cell types. Finally, t-SNE clustering of single-cell glioma RNA sequencing data with single-cell organoid data revealed transcriptional similarities between H3K27M and pre-EMT neural stem cells. In conclusion, we observed aberrant activity of the EMT in H3K27M gliomas. Our data suggest that the H3K27M mutation is associated with a pre-EMT cell phenotype, and that this mutation may cause EMT arrest or de-differentiation. Citation Format: Allison R. Cheney, Lauren M. Sanders, Lucas Seninge, Holly C. Beale, Ellen Towle Kephart, Jacob Pfeil, Katrina Learned, A. Geoffrey Lyle, Isabel Bjork, David Haussler, Sofie R. Salama, Olena M. Vaske. Candidate differentiation stall in epithelial mesenchymal transition in H3K27M diffuse midline glioma [abstract]. In: Proceedings of the AACR Special Conference on the Advances in Pediatric Cancer Research; 2019 Sep 17-20; Montreal, QC, Canada. Philadelphia (PA): AACR; Cancer Res 2020;80(14 Suppl):Abstract nr B06.